Most underperforming Google Ads accounts aren't broken in some exotic way. They're leaking money in four or five predictable places, and nobody has looked properly in a while.
An audit is just looking properly, in a fixed order, so you don't get three clicks deep into one campaign and forget the rest.
This is the checklist I work through on every account. It takes about an hour the first time. You don't need any paid tools — everything here is visible inside Google Ads, Google Analytics and your own website. Download the free Google Ads Audit Tracker to work through it alongside the guide.
Before you start
Two things make the difference between an audit and a browse.
Set the date range wide. Use the last 90 days, or the last 12 months for an account with low volume. A 7-day view will send you chasing noise.
Write findings down without fixing them. The temptation is to fix each problem as you find it. Don't. You'll lose your place, and worse, you'll make five changes at once and never know which one mattered. Collect everything first, then prioritise.
Open a blank document with two columns: what I found, and what it's costing.
Free Google Ads Audit Tracker. A pre-built spreadsheet with all 15 checks, where to find each one, status and priority dropdowns, a search terms working sheet that totals your wasted spend, and a progress summary. Ready to fill in. Download the template (.xlsx)
Measurement first (checks 1–4)
If the numbers are wrong, every decision you make afterwards is wrong with confidence. This is always where I start, and it's where I find the most damage.
1. Are conversions actually firing?
Go to Goals → Conversions → Summary. Look at the "Conversions" column over the last 30 days.
If a conversion action shows zero, it's broken or it's not happening — and you need to know which. If one shows a suspiciously round pattern, it's probably double-firing.
The fastest sanity check: submit your own contact form, wait a few hours, and see whether it appears. Half the accounts I look at have at least one conversion action quietly recording nothing.
2. Are you counting the same conversion twice?
Common causes: the tag is on both the form page and the thank-you page, or a conversion is being imported from Analytics and tracked directly by the Google Ads tag.
Check the Source column on your conversion actions. If you see the same event arriving from both "Website" and "Google Analytics (GA4)", you're inflating your own results and bidding on fiction.
3. What's actually in the "Conversions" column?
This one catches people out. Each conversion action has a setting for whether it counts toward the "Conversions" column that bidding optimises against — Google calls these primary actions — or whether it's just observed.
Newsletter signups, PDF downloads and page views marked as primary will drag automated bidding toward cheap, low-value actions. Enquiries and sales should be primary. Everything else should be secondary.
I've seen accounts where the only primary conversion was "viewed the contact page." The bidding was working perfectly. It was just optimising toward the wrong thing.
4. Attribution model and conversion window
Under each conversion action, check the conversion window — how long after a click a conversion still counts.
A 30-day window on a business with a two-day decision cycle overstates performance. A 7-day window on a service with a six-week sales cycle hides conversions that genuinely happened, and starves the campaigns that earned them.
Match the window to how long your customers actually take to decide.
Where the money goes (checks 5–8)
Now that the numbers mean something, follow the spend.
5. The search terms report
Campaigns → Insights and reports → Search terms. Sort by cost, descending.
This is the single most valuable screen in Google Ads and the most neglected. You're looking for terms you're paying for that you'd never have chosen:
- job seekers ("[your service] jobs", "[your service] salary")
- people looking for free versions ("free", "DIY", "how to do it yourself")
- students and researchers ("what is", "definition", "examples")
- competitor brand names you don't want to fight over
- the wrong service entirely
Every one of those with meaningful spend and no conversions is money you can stop spending this afternoon.
6. Negative keyword lists
Check Tools → Shared library → Negative keyword lists, then check whether those lists are actually applied to your campaigns. Creating a list and never attaching it is one of the most common mistakes in the platform.
While you're there, look for the opposite problem: over-aggressive negatives blocking terms you want. A broad negative like "cheap" seems sensible until you notice it's blocking "cheap [your service] vs expensive."
7. Where the budget sits versus where the results are
Pull up your campaigns, add the Search lost IS (budget) and Search lost IS (rank) columns.
- High lost IS (budget) on a campaign that converts well means you're capping your best performer. Feed it.
- High lost IS (rank) means your bids or quality are too low to compete — a different problem with a different fix.
- Steady spend on a campaign with no conversions over 90 days is a decision you're making by not making it.
Budget should follow opportunity, not history. Most accounts are funded according to what someone decided eighteen months ago.
8. Auto-applied recommendations
Recommendations → Auto-apply. Check what's switched on.
Google will happily add broad-match keywords, rewrite your ad copy and raise budgets on your behalf if these are enabled. Some are harmless. "Add broad match keywords" and "Use optimised ad rotation" deserve a deliberate decision rather than a default.
If your account changed behaviour suddenly and nobody touched it, this is usually where the answer is.
Targeting and intent (checks 9–11)
9. Match types and keyword bloat
Look at your keyword list. Two failure modes:
Too broad — an account that's almost entirely broad match with no negative discipline is an account funding Google's experiments. Broad match works, but only with solid conversion tracking and a maintained negative list, which is why checks 1–6 come first.
Too many — 400 keywords across six ad groups means nothing accumulates enough data to optimise. Tight groups of closely related terms beat sprawling lists.
If you've got the same keyword sitting in several ad groups, you're bidding against yourself. That's the paid-media version of keyword cannibalisation, and it has the same effect: splitting signal that should be concentrated.
10. Location targeting — check the setting, not just the map
This is the check that most often finds live money, and it's buried.
In each campaign: Settings → Locations → Location options. There are two settings, and the default is not the one most small businesses want:
- Presence: People in or regularly in your targeted locations — what you almost certainly intend.
- Presence or interest: People in, regularly in, or who've shown interest in your targeted locations — the historical default, which serves your ads to people anywhere in the world who searched something related to your city.
If you've targeted Auckland and you're getting clicks from outside New Zealand, this setting is why. Switch to "Presence" and a chunk of waste disappears immediately.
Also check the exclusions tab. Targeting a region while forgetting to exclude the areas you can't service is a slower version of the same leak.
11. Audiences and exclusions
Check whether you're excluding people who've already converted. Continuing to pay to reach existing customers through acquisition campaigns is common and entirely avoidable.
Then check your Search Partners and Display Network expansion settings on search campaigns. Display expansion on a search campaign quietly moves budget into a completely different kind of placement with completely different intent. For most small accounts, turning it off makes the numbers make sense again.
Ads and destination (checks 12–14)
12. Responsive search ad coverage
For each ad group, check you have at least one responsive search ad, and look at the asset view to see which headlines are marked Low, Good or Best.
You're looking for three things: enough headlines to give the system room, headlines that say genuinely different things rather than fifteen rewordings of your business name, and at least one that names the specific problem the searcher has.
Ad Strength is a rough guide, not a score to chase. An ad rated "Average" that converts beats an "Excellent" one that doesn't.
13. Assets (the things formerly called extensions)
Sitelinks, callouts, structured snippets, call assets, location assets.
These are free real estate. They make your ad physically larger, push competitors down the page and let you answer objections before the click. An account running ads with no assets is leaving the easiest available win on the table.
Check they're present, and check they're not out of date — a sitelink pointing at last year's offer is worse than no sitelink.
14. Does the landing page match the ad?
Click your own ads. All of them.
The question isn't "is this page nice." It's whether someone who searched that exact term, read that exact ad, and landed on that exact page finds what they were promised above the fold.
Sending all traffic to the homepage is the classic version of this. The ad makes a specific promise; the homepage makes a general one; the visitor does the reconciling, and often doesn't bother.
While you're there, check the page actually loads quickly on a phone — a slow landing page burns budget on every single click, and it's worth running the same Core Web Vitals checks on wherever your ads point. If the traffic is arriving but the enquiries aren't, the problem is often further down the page than you think, which I've gone through in traffic but no enquiries.
Structure (check 15)
15. Does the account structure make sense to a stranger?
Last, step back and look at the campaign list as though you'd never seen the account.
Can you tell what each campaign is for from its name? Are campaigns split in a way that matches how the business actually makes money — by service line, by margin, by location? Or has it grown by accretion, with campaigns added whenever someone had an idea and never removed?
Structure isn't cosmetic. It determines what you can control separately: budget, targeting, bidding. If two services with wildly different margins share a campaign, you can't fund them differently, which means you're funding them wrongly.
An audit isn't about finding faults for the sake of it. It's about finding the handful of places where a small change moves real money.
The 30-minute version
If you genuinely only have half an hour, do these five in this order. They find most of the money:
- Conversion tracking is firing and counting the right actions (checks 1 and 3)
- Search terms report, sorted by cost (check 5)
- Location targeting set to "Presence" (check 10)
- Search Partners and Display expansion off on search campaigns (check 11)
- Click your own ads and check the landing pages (check 14)
These five are flagged in the tracker's "In 30-min audit" column, so you can filter to them and ignore the rest.
What to do with the list
You'll finish with more findings than you can action at once. Sort them by two questions: how much is this costing, and how quickly can I fix it.
Do the free, fast, high-cost ones first — negatives, location settings, display expansion. Those are usually same-day changes with same-week savings.
Then fix measurement properly, because everything you do afterwards depends on it.
Then, and only then, start making bidding and budget decisions. Changing bids on top of broken tracking is how accounts end up worse after an audit than before one.
One more thing worth saying: an audit tells you whether the account is running well. It doesn't tell you whether paid search is the right place for the budget in the first place. That's a separate question, and I've worked through it in SEO vs Google Ads for NZ small business. If you're running paid social alongside search, the same setup order applies there, and it's covered in Meta Ads for small business.
Download the free Google Ads Audit Tracker (.xlsx) and run your first audit this week. It takes an hour, and it's the cheapest hour in your marketing calendar.
FAQ
Is there a free Google Ads audit template?
Yes — the Google Ads Audit Tracker above is a free .xlsx spreadsheet covering all 15 checks, with the menu path for each one, status and priority dropdowns, a separate search terms sheet that totals the spend you've marked for negatives, and a progress summary. It opens in Excel, Numbers or Google Sheets, and there's nothing to sign up for.
How often should I audit my Google Ads account?
A full audit once a quarter is enough for most small accounts, with a quick monthly look at the search terms report and conversion tracking. Audit immediately if performance changes suddenly, if someone new takes over the account, or before you increase budget — scaling an account with an undiagnosed leak just scales the leak.
How long does a Google Ads audit take?
About an hour for a small account using this checklist, and longer for accounts with many campaigns. The abbreviated five-check version above takes roughly 30 minutes and finds the majority of obvious waste.
Can I audit my own Google Ads account without an agency?
Yes. Everything in this checklist is visible in the standard Google Ads interface with no paid tools. You need administrative or standard access to the account, not just read-only, if you want to change settings once you've found the problems.
What's the most common Google Ads mistake in small accounts?
Broken or misconfigured conversion tracking, because it invalidates everything downstream — automated bidding optimises toward whatever you tell it is valuable, so an account counting page views as conversions will efficiently buy page views. After that, the two most common are location targeting left on "Presence or interest" and the search terms report never being checked.
What does "Search lost IS (budget)" mean?
It's the share of available impressions your ads didn't receive because your daily budget ran out. A high figure on a campaign that converts profitably means you're capping your best performer — the demand is there and you're choosing not to serve it.
Should I use broad match keywords?
Broad match can work well, but only on top of accurate conversion tracking and a maintained negative keyword list. Without those two things it tends to spend into loosely related searches. If you're unsure whether your tracking is solid, start with phrase and exact match, and expand once checks 1–4 come back clean.
Jayne Hamilton
Digital marketing strategist working across SEO, paid media, analytics and website optimisation.
More about me